===================================================================
REPLICATION MATERIALS
===================================================================

Title: Modeling Issue Competence Over Time:    A Bayesian Framework for Estimating Dynamic Issue Ownership
Author: Dominic Nyhuis, Jona-Frederik Baumert, Jeongho Choi, Sebastian Block, and Morten Harmening
Correspondence: Jeongho Choi (j.choi@ipw.uni-hannover.de)
Date: December 2025

===================================================================
Overview
===================================================================

This replication package contains all data and code necessary to reproduce the figures and tables in the main text and appendix of the manuscript.

===================================================================
SOFTWARE REQUIREMENTS
===================================================================

Software
	- R version 4.5.1 (2025-06-13) -- "Great Square Root"

================================================================================
DATA FILES
================================================================================

The replication package includes the following datasets:

1, issue_ownership_rev_v2.Rdata
	- German Longitudinal Election Study data on Parties’ issue competences
	- Raw data (Complete list of datasets of GLES surveys is included Table A1)
	- Used for estimating the following Bayesian models
		- full_base_rev_v2.Rdata 
	        - full_base_pid_rev_v2.Rdata
		- full_topic_rev_v2.Rdata
                - full_topic_pid_rev_v2.Rdata
                - full_topic_biannual_interact_pid_rev_v2.Rdata
	- Used for Figure C1
	- Used for Figure C2

2. full_base_rev_v2.Rdata: Static Macro-competence Model
	- Model 1 in Table 1

3. full_base_pid_rev_v2.Rdata: Static Macro-competence Model with party intercepts
	- Model 2 in Table 1

4. full_topic_rev_v2.Rdata: Static Macro-competence Model with issue intercept
	- Model 3 in Table 1
	- Used for Table A3

5. full_topic_pid_rev_v2.Rdata: Static Macro-competence Model with party and issue intercepts
	- Model 4 in Table 1
	- Used for Figure 1
	- Used for Table A3
	- Used for Table A4
	- Used for Figure A1

6. full_topic_biannual_interact_pid_rev_v2.Rdata: Dynamic Issue Competence model
	- Used for Figure 2
	- Used for Figure 3
	- Table A2
	- Used for Figure C3

================================================================================
CODE FILES
================================================================================

1, Replication_Analysis_Issue_Competence.R
	- Reproduce all figures and tables for the manuscript and appendix

2. Replication_Model Estimation_Issue_Competence.R
	- Reproduce Bayesian Models 

================================================================================
INSTRUCTIONS FOR REPLICATION
================================================================================

IMPORTANT: All data files must be in the same folder as the do-file.

1. Download all files to a single directory on your computer

2. Open R and set your working directory to the folder containing all files:

	- In R, use the File pane: File > More > Go to the folder that contains all files > Set as Working Directory 
      	- Or type in R: setwd(Replace with your actual folder path)
      
3. Open Replication_Model Estimation_Issue_Competence.R
	- This step is unnecessary to reproduce the results for the manuscript and the appendix
	- since all estimated models are included in the replication folder.
	- So, if you want to reproduce the main result, you can skip this process
	
	- If you want to reproduce the Bayesian modeling on your own, you can run this code.
	- However, reviewers should modify the code by increasing the number of iterations
	- to the full value stated in the paper to obtain the correct results. 
	- In the replication code, we set the number of iterations to 500 to avoid excessive computation time

4. Open Replication_Analysis_Issue_Competence.R and run the code to reproduce all results of the main text and the appendix

5. The log file "replication_log.txt" will be created automatically in your working directory and will contain all output

